Abstract
Background:
In many resource-poor settings such as Peru, children affected by HIV have a high prevalence of neurodevelopmental delays (NDDs) and remain excluded from adequate treatment.
Methods:
Community health workers (CHWs) administered NDD screening instruments to assess child development and associated caregiver and household factors in 14 HIV-affected parent–child dyads. Focus group discussion with caregivers was conducted to explore their needs and behaviors around early child stimulation and to assess their perceptions of the screening experience.
Results:
Over 70% of the children had abnormal classification in at least 1 (out of 5) developmental domains according to Ages and States Questionnaire–provided cutoff scores. Caregiver depression and stress were associated with abnormal development as were some parenting behavior factors. Knowledge about child development was low. Caregivers felt testing and discussing results with a CHW were very insightful. Reported caregiver behavior differed between caregivers with HIV-infected children and those with uninfected children.
Conclusion:
Taken together, these exploratory quantitative data suggest that parenting behaviors associated with low child development scores may be modifiable and that community-based testing is well received and informative to these HIV-infected caregivers.
Introduction
Globally, an estimated 200 million children are at risk for neurodevelopmental delay (NDD). 1 The prevalence and impact of NDDs are disproportionally greater in resource-limited countries as well as in low-resource settings within wealthier countries. Neurodevelopmental delay risk among children affected by HIV (ie, children born to an HIV-positive parent) is magnified by the cycle of extreme poverty, malnutrition, lack of access to health services as well as the neurotropic impact of the HIV virus among those children who are infected. In addition to immunosuppression and poor nutrition, 2,3 many socioeconomic factors predict NDD risk among HIV-affected children, including low education, 4,5 family conflict, 4,6 unstable home environment, 6,7 lack of caregiver consistency, 8,9 and poor parenting quality. 10 –13 Thus, developmental delays among HIV-affected children should be contextualized within the family and home environment. 14
Despite the wealth of evidence that HIV-affected children are at greater risk for NDD than unaffected children, 15 –18 in many developing countries only extreme cases of NDD are addressed. Even when NDD is recognized, impoverished families who are fighting to survive with comorbidities of HIV, poverty, stigmatization, and other psychosocial stressors may be unable to prioritize their children’s healthy neurocognitive development as a result of lack of knowledge and/or lack of access to services. Furthermore, among HIV-infected children, providers and parents may attribute NDDs to biological factors and, therefore, are less likely to pursue interventions focused on neurocognitive stimulation and environmental enhancements.
Despite the magnitude of the problem, few studies have explored associations with caregiver behavior and well-being. 19 Through our experience in Peru, providing outreach to families affected by HIV, 20 our local team of community health workers (CHWs) identified the urgency in addressing the high rates of NDDs among HIV-affected children. Before designing an effective community-based intervention, we conducted a pilot in which we administered NDD, home environment, and psychosocial measurement instruments to 14 mother–child dyads comprising an HIV-infected mother and a child between 1 and 3 years of age. Our goals of the pilot were to (1) explore NDD among HIV-affected children and (2) understand how parenting behavior can impact delay in this cohort. We used quantitative findings to begin to understand the areas of NDD among the children and qualitative methods to develop greater insight and solutions from family members themselves that would, in turn, inform the intervention.
Methods
Study Team
Two specialists from the United States in pediatric neuropsychology (CV, SL) worked with the Peru team to carry out the activities described below, using community-based participatory methods. One of the specialists (CV) focused on cultural considerations of neuropsychological assessments in Spanish-speaking populations. The Peruvian team consisted of a nurse and 4 CHWs who were selected based on prior clinical and research experience working with HIV-affected families and with children.
Selection of Study Instruments
No instruments in Peru were available that met study needs, therefore, instruments were selected by US psychologists with experience in international assessment and included the Ages and States Questionnaire–Third Edition (ASQ-3) 21 and the Home Observation for Measurement of the Environment (HOME). 22 The ASQ-3 is a parent-completed developmental screening tool used to monitor whether a child is making appropriate developmental progress by asking parents to report the achievement of specific observable skills or behaviors using structured questions. The ASQ-3 is for children between the ages of 1 month and 5½ years and is comprised of 21 developmental intervals that each contain 30 items that are divided into 5 developmental domains including communication, fine motor, gross motor, problem solving, and personal social. The ASQ-3 is the most widely used screening tool in child care in the United States 23 and in international research studies, 24 and research was underway to norm the ASQ in Peru at the time this study took place. Administrators are often trained laypeople, it is available in Spanish, 25 –27 and it can be administered in a relatively brief period of time (approximately 10-15 minutes). 28 Cutoffs were established in the United States, and while they are not applicable in Peru, they were useful in this study to measure risk compared to similar contexts and gauge difference between domains. The use of the more comprehensive measures such as Bayley Scales of Infant and Toddler Development–Third Edition was discarded even though it has been administered in Peru, because the administration time and expertise in psychology required to administer the test were not considered feasible for this setting. 29 The Home Observation for Measurement of the Environment is a complementary assessment that uses both the home-based observation of the parent and child and a parent questionnaire to assess the quality and quantity of stimulation and support available to a child in the home environment, with the focus being on the child in his or her environment and the child as a recipient from various inputs (eg, objects, events, and transaction) occurring in connection with the family surroundings. 22,30,31 The HOME can help capture the family system organization, routines, and involvement with extended family and the larger community, all impacting the child’s development. The assessment involves a home visit to observe the child and the child’s primary caregiver but may also capture other members of the household. There are 4 versions of the inventory based on the age of the child, each containing specific scales. This population received the infant/toddler version with the following subscales: (1) responsiveness, (2) acceptance (lack of punitiveness), (3) organization, (4) play/learning materials, (5) involvement (stimulation of development), and (6) variety of experience. Both the ASQ-3 and the HOME are frequently used in research in resource-limited settings 31 and are beginning to be used in similar settings in South America. 27,32,33
In addition, we selected the Hopkins Symptom Checklist, 34,35 Global Stress Score, 36 and the Duke University of North Carolina Social Support 37,38 to assess the psychosocial status of the primary caregiver (ie, mother); these psychosocial instruments had been previously validated in Spanish 39 –42 and the Peru team had prior experience administering them to HIV-positive adults. The CHWs administered all tests and questionnaires as well as a brief intake form with demographic and health information on both the parent and the child.
Items were reviewed for cultural appropriateness, and modified items were piloted according to the accepted practice. 43
Selection and Enrollment
We prospectively enrolled 14 children aged 1 to 3 (7 HIV positive and 7 HIV negative) and their HIV-positive mothers. Inclusion criteria for the child were (1) no known diagnosis of NDD and (2) born to a sero-positive mother. The inclusion criteria for the mother were (1) HIV-positive status and (2) primary caregiver of the child. Participants were recruited from families that had previously received outreach support from Socios En Salud. The study team approached mothers by telephone or home visit, providing an explanation of the study and arranging a follow-up home visit at a time convenient to the mother (eg, weekend for a working mother). At the follow-up home visit, the CHW explained the purpose of the study to the mother who provided informed consent for herself and her child. Approval for the study was obtained from the Hospital del Niño in Lima and Partners Healthcare in Boston. Each family was given educational toys (Lego set and age-appropriate books) as well as a food basket.
Home Encounters
Access to patient homes in outlying poor neighborhoods of Lima was challenging because homes are frequently built on steep mountains that are only accessible via motor taxi from the main road and then by foot, climbing up rocky slopes using hands and feet. The team encountered a variety of challenges in applying the instruments at home but did their best to adapt successfully to each situation. Some mothers were not at home when CHWs arrived, others were initially distrustful. Many patient homes did not have chairs or tables and lacked space for children to walk or run. In one case, a child performed drawing activities on an upside down bucket and in another situation a CHW brought a hard folder on top of a chair as a makeshift table. To this end, CHWs adjusted by using other items (eg, bricks or the side of the bed) when chairs or tables were unavailable. There were also houses inhabited by multiple families, overcrowded living spaces, frequent interruptions, and little privacy. As a result of these conditions, mothers were sometimes uncomfortable talking about having HIV. In one case, the sister of the patient did not know she had HIV, and thus, the team suggested that the mother ask her to play with the child outside so she could talk openly. Additional cases required the CHWs to return the following week to finish evaluations when children got too tired. CHWs performed chart reviews for clinical data in the hospital corresponding to their zone, since participant recall was variable.
The team found that the ASQ took about 25 minutes to administer, rather than the 10 to 15 minutes mentioned previously. The duration depended on the mother’s educational level, as the team sometimes needed to further explain questions to ensure comprehension. The HOME took between 45 and 60 minutes to administer, slightly shorter than the 45 to 90 minutes reported in the literature.
A final home visit was conducted with each family to give them the food basket and discuss the results of their child’s development. Recommendations for stimulation activities were given according to development domain. General recommendations were given to all children, and specific ones were emphasized in the specific area(s) of delay.
Upon completion of the pilot, a focus group with 8 mother participants was conducted to solicit participants’ reactions and feedback about the encounters. The focus group, led by the study nurse, was audiotaped and a scribe took notes.
Analysis
Data were entered into Microsoft Excel and checked for completeness and quality. Missing clinical data were sought by chart audit. All ASQ-3 and HOME data were complete. Analysis was performed in SAS (version 9), including calculation of means, standard deviations, and percentages as well as tests of association using the Fisher test and t tests. Findings were presented back to CHWs and psychologists and health professionals from the United States and the Hospital del Niño in Lima.
For qualitative analysis, recordings of focus group discussions were transcribed by an individual fluent in Spanish. Interview transcripts were coded inductively with content analysis, using the key variables identified during the pilot study. Themes were then integrated into a theoretical scheme, which was reviewed by the Peru study team to triangulate with their observations and quantitative findings.
Results
Quantitative Findings
Based on the ASQ-3, 10 (71.4%) children had an abnormal classification in at least 1 domain. As shown in Table 1, caregiver depression and stress were associated with abnormal ASQ-3 (P < .2).
Cohort Characteristics Associated with Abnormal ASQ.a,b
Abbreviations: ASQ, Ages and States Questionnaire; HAART, highly active antiretroviral therapy; HOME, Home Observation for Measurement of the Environment; NDD, neurodevelopmental delay; SD, standard deviation.
aN = 14.
bLow educational status defined as did not enter high school.
c P value < .2.
d P value < .1.
Although no cutoff points are specified in the HOME manual, recommended interpretation suggests a home environment may pose risk to some aspect of the child’s development if the subscore falls in the lowest quartile of the score range; more importantly, HOME scores are to be interpreted based on the observation of the home environment. 22 In our cohort, the domains with the lowest scores were related to parenting behavior (percentage at or below the lowest quartile) including ability to respond (21.4%), acceptance (21.4%), and involvement (21.4%); the same areas were found to have a negative impact on early child development (ability to respond, P = .1 and acceptance, P = .2). In the HOME manual, responsivity is defined as “the extent to which the parent responds to the child’s behavior including reinforcement for desired behavior and communication through words and actions” and acceptance is defined as “parental acceptance of the child’s behavior that is less than optimal and examines avoidance of undue restriction and punishment.” 22
These findings were confirmed by the study team who felt that the physical environment, while limited, posed less impact on the child’s potential stimulation than deficiencies observed in the parent’s behavior, either due to lack of knowledge or due to the inability to prioritize stimulating child interactions because of competing psychosocial and economic stressors.
Qualitative Findings
Overall, mothers expressed a positive response to the home encounter. They thought home-based, face-to-face meetings to evaluate for NDD were preferable to hospital- or clinic-based assessments, because it allowed the evaluator to understand the reality that the mother faced in terms of her home environment and socioeconomic status. They were also appreciative that the encounter was administered by a CHW who were perceived as trusted members of the participants’ community. As one woman said, “the health promoters are people who help us one hundred percent, they are trained, they stand by our side, and we have never heard of any case where they have told someone about our diagnosis.” The home visits were well received despite initial concerns about confidentiality. However, in one case, the patient manifested after the home encounter that she wished that the evaluation had taken place outside of the home. In this situation, the CHW suspected interpersonal conflict with the mother’s partner. For this reason, CHWs believe mothers should be asked about their preference of location of assessment meetings in the future.
Participants were then asked for specific feedback about the experience with the evaluation tools (eg, questions that were difficult to understand or answer, difficulty maintaining concentration throughout the interview while caring for child, feelings of discomfort or embarrassment, etc). Their responses were uniformly positive. They did mention, however, that their children did not adapt easily to new environments and that future visits in the homes would give them a rare chance to speak openly about their concerns and emotions regarding their child’s development.
Mothers demonstrated fairly low knowledge of NDD and positive parenting behaviors. Many mothers were surprised when their child was diagnosed with a delay, stating that they had viewed those behaviors as “normal” or “natural,” as mentioned by a study participant, “I think it is normal for a child that is 2 or 3 years old to not speak correctly.” They were also surprised to learn that caresses or positive words were especially important for their infant and that they should encourage their child to talk. Others did not feel they were able to stimulate their child because of their own perceived lack of education.
There was a difference in attitudes among mothers with HIV-positive children versus HIV-negative children. All mothers with HIV-positive children felt profound guilt for passing on the infection to their child. One mother commented, “When I found out that I had this disease, and my child did too, I wanted to kill myself, because I thought, ‘I made him have this.’” Mothers of HIV-positive children also tended to overprotect their children, isolating their children or limiting their child’s activities for fear of them getting sick. These mothers expressed hope that their child would be adherent to their medications and that they would “forgive” their mothers. A mother talked about her dreams for her children, “My dream for my child is that his brother will be a doctor, so he can take care of him.” They also expressed concern and anxiety about their own ability to care for their sick child in the future.
Unexpectedly, most mothers in the focus group reported they were appreciative of the opportunity to discuss their experiences with other women who faced similar challenges. Women mentioned they did not have other people in their social environment to talk about HIV-related difficulties and described a sense of support through sharing observations during the focus group. “[In the neighborhood where I live], they are not like us, they are afraid of the disease and they will always keep their distance. [Here], we can learn from each other.” After closure of the formal focus group discussion, the team continued the discussion to provide support and validation to the women, to ease parents’ expressed feelings of guilt and depression, to offer guidance about child rearing in the context of HIV, and, finally, to explore the relative benefits of accepting versus denying their own diagnosis. All women expressed a need for ongoing group support for mothers living with HIV.
Discussion
This report describes a pilot experience in which CHWs administered instruments to assess NDD risk and related environmental factors in the homes of HIV-affected children in Lima, Peru. The deployment of CHWs to administer these instruments in the homes, as well as their ability to adapt to the challenges posed in the field, enabled a thoughtful evaluation of the child’s developmental status within the larger context of caregiver and household influences. The assessment also provided an opportunity to explore parents’ understanding of the importance of neurocognitive stimulation in a way that was perceived as supportive and nonjudgmental to the mothers. Our exploratory quantitative data suggest that the risk of NDD is high in this HIV-affected cohort. Compared to a similar study among 129 HIV-unaffected Peruvian children from all income levels, the magnitude of delay among our cohort is almost 2-fold (71.4% compared to 38.7%). 32
While this cohort is too small to make definitive conclusions from our quantitative results, they are supported by our qualitative findings, which provide deeper insight into the challenges faced by the caregivers of these children. Mothers reported feeling depressed, isolated, and stressed because of their own HIV status, and further, they experienced guilt, stress, and anxiety related to their role as mothers. Much of the anxiety expressed by mothers related to their uncertainty about their own future, their ability to care for their children, and concerns about the future health and medication adherence of their HIV-infected children. Such psychosocial stressors could result in less responsiveness and involvement as parents and potentially greater isolation of the child. Some mothers felt tremendous guilt and fear of future loss that they could not permit their children to be actively independent outside of their own supervision. In other cases, mothers had trouble emotionally connecting with their children. Literature supports these findings, suggesting negative outcomes to parental depression in low-resource settings such as behavioral difficulties, attachment problems, 44 –46 depression in children up to 16 years of age, 46 and poor cognitive functioning. 1,47 Mothers with HIV whose child has also been diagnosed with HIV may be particularly important to target for parenting behavior and psychosocial support during this critical period.
The positive response to home-based assessment contrasted with feelings about the hospital or clinic, where mothers said their child’s development was rarely assessed or discussed because of the focus on HIV management. Furthermore, mothers sometimes manifested feeling devalued or stigmatized at health establishments. Community-based strategies to deliver simple, evidence-based NDD screening and treatment may offer an efficient and acceptable means to reach the most at-risk children and provide emotional support and validation to their caregivers.
We were unable to correlate the ASQ-3 with other instruments, due to the small sample size and time restrictions. Future work could focus efforts on norming and validating early child development screeners such as the ASQ-3 to local resource-poor settings and correlating them with local and internationally validated instruments.
This study was not primarily an academic exercise; priority was given to learning about local cultural context in order to best inform how to effectively deploy assessment instruments and design community-based interventions to address the needs of a specific population. Unexpectedly, we found that the delivery of the assessments tools, particularly the ASQ-3 and HOME instruments as well as participation in the focus groups, sparked useful conversations about child development with families. We believe these instruments could help open discussions with parents about child development, particularly if delivered by a CHW—a well-trained person from the same community who can intuitively understand the family’s social and economic context. Future work is needed to develop and test a screening and intervention “package” delivered by CHWs to address developmental needs of HIV-affected children.
Footnotes
Acknowledgments
The authors would like to thank Gianina Jeri for her work supervising health promoters’ training and field workers during this study.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the David Rockefeller Center for Latin American Studies (DRCLAS) at Harvard University under grant name “Community-based Assessment and Intervention for Developmental Delay in HIV-affected Children”.